The objective is to continue current research to describe the natural Alzheimer's disease (AD), to describe the borders between normal aging and AD; to examine the stage-specific prognosis of mild, moderate, moderately severe, and severe dementia; to determine factors associated with a relatively malignant and a relatively benign course, and to identify possible prognostic subgroups and markers. In an initial longitudinal study (T1), 176 subjects 60 to 84 years of age were followed over a mean interval of - 4 years. Nine-year mean interval follow-up on these subjects has now been completed. In a subsequent ongoing study (T2), 363 subjects seen at baseline are being followed over - 4 years. Of this T2 cohort, follow-up has been completed on 275 subjects seen initially prior to 1987 and follow-up is presently being completed on the 88 subjects from the 1987 T2 cohort. Additionally, 265 subjects have been seen from 1988 until the present who are now eligible for follow-up. Loss to follow-up has been less than 2% per annum across all of the above completed studies. In part, this excellent success rate over resent intervals up to 9 years is due to: (a) our excellent relationship with our subjects and their families, and (b) our determination to follow our subjects into their residential and nursing home settings when necessary. Important results from these studies at the present time include: (1) greatly improved definition of the borders and boundaries of clinically normal aging and dementia (especially, AD); (2) greatly improved measures for tracking the course of AD and for assessing the clinical phenomenology of AD, and (3) initial validation of a hypothesized detailed estimation of the precise temporal course of AD. We now propose to follow-up the previously studied and followed cohorts after a further five-year interval, 14 years and 9 years after their initial baseline assessment for T1 (N - 176) and T2 (N- 363), respectively. Simultaneously, we propose to complete an initial follow-up (- 4 years) on more recently studied subjects (T3, N - 265). We will continue to conduct home visits and nursing home visits when necessary. Postmortem studies will continue to be conducted. The goals are to confirm and extend the findings of our initial studies and to investigate the hypotheses raised by our findings to date. The results will improve our ability to counsel patients and families of AD victims with respect to prognosis at all stages of the illness, to differentiate the clinical and behavioral borders of aging/AD and to detect incipient AD.